A Useful Grid-Bias Unit

Radio Record, Volume II, Issue 52, 12 July 1929, Page 34

A Useful Grid-Bias Unit

Gives Any Required Voltage

B

MEGOHM

ECAUSH the grids of the valves are the most sensitive parts of a receiving circuit, the application of a bias potential giving the slightest trace of hum from mains, is not permissible, The earlier the stage in the receiver, the more this applies since subsequent amplification shows up the slightest defect. It is only with grid bias for audio amplifiers that we are particularly concerned, and this article will describe an efficient unit that may be constructed and used to provide two or three bias voltages, each varlable from zero to one hundred volts. The writer feels that this unit will suit many constructors who have an eliminator in use, and wish to obtain a high-bias voltage without in any way altering the "B" eliminator or reducing its voltage, and without introducing any of the difficulties liable to be encountered in obtaining bias from the voltage drop in the eliminator resistance. Some may question the desirability of having a separate unit with its attendant consumption of current, but as the total current taken from the mains is only from 6 to 10 watts, little consideration is required for this item. Actually the unit is a half-wave "B" eliminator with certain alterations to adapt it for "C" supply. The dif-° ference necessary to make a "B" unit will be described so that any constructors who wish to construct an eliminator for a crystal amplifier may do so. The points common to both units will be first described, the difference, chiefly in the resistance and condenser connections, being dealt with afterwards.

It should be stated here that this unit may be incorporated in a "B" eliminator by anybody about to build one, in which case instead of the separate transformer, there would be an extra filament winding and a bias winding to give 150 volts, provided on the main transformer. The Transformer. THE transformer is constructed of either lin. stalloy strips, or the equivalent stampings that have already been mentioned in these columns. For the stampings, the core is built to 1tin., and for strips to full one inch. ~ The primary winding is of 32's s.w.g. enamelled wire, 1800 turns on the stampings, and 1850 on inch stalloy. A wooden former of the required seetion is to be made, not less than three inches long, and arranged to turn on a spindle with winding handle. The stampings are built to 14in. thickness and no spool will be required. Strips of increased thin brown paper must be cut, bare 2}in. wide. These are used between layers of wire. As a preliminary, a layer of thin card, 2hin. wide, is secured round the former and covered with adhesive tape, the edges overlapping. Now a layer of the: brown paper, and the primary winding, occupying 11 layers, may be put on. Leave a space of jin. at each end of the layer of wire, and be careful to wind the turns evenly without crossing the wires one over another. Cover the primary with a layer of paper, then a layer of tape, and another of paper.

The filament winding, of 24’s, d.c.c, comes next, occupying less than one layer. If 22’s is used and wound witbout loss of space, the 46 turns will just go in one layer. Tap at ‘24th turn for ‘the centre tap-though not the centre, it will be correct for rectifying purposes. Tap at 33 turns for a 4-volt and 40 turns for a 5-volt valve. Thése taps will give sufficient filament heating for the emission required, and give very long life to the rectifying valve. For a "B" eliminator, add 4 turns, making the total 50, and make each filament tap 2 turns higher. Cover this winding with a layer of tape and then one of paper. The secondary winding, of 36's, s.w.z., enamelled wire follows, and for the bias unit 1390 turns are required, giving 150 volts, of which 50 volts will be dropped in the rectifying valve and choke. For a "B" eliminator 1850 turns will be required on the secondary, giving about 150 volts when rectified and smoothed. The clamps of this small transformer may be of wood, 1 by 3-8 on stalloy strips, but may be slightly thicker and narrower on the stampings, and placed on the two ends parallel to the centre core. The "Listeners’ Guide" gives complete details of transformer construction, and will be found handy for reference. A fuse panel must not be omitted, and may be of fibre or ebonite, measuring about 13 by 4 inches, supported above the transformer by a shaped piece of sheet brass scrgwed to the side of the clamps. The fuses for all

positions may be strips of tinfoil 1-16 inch wide, or narrower on the secondary negative. Only one fuse is required for the half-wave secondary. Its function is to save re-winding the secondary coil in case of an accidental short-circuit in that portion of the circuit before the rectifier. (as ae. ar am

Threaded screws for the clamps will be 2}in. long, or a trifle less. Twenty oft. lengths of lin. stalloy will be used. Shellac the strips well before cutting up. The long piece of stalloy will be 3 3-8in. and the short ijin. Cut 128 of each. When the winding is removed from the former, the ends are to be filled in with the black composition from the tops of old batteries or elsewhere. Melt the composition in a small seamless tin box or lid and pour on gradually, smoothing with a knife. Bring out leads in the most convenient positions, when possible-pri-maries both at one end, with ali filament leads except that from -the last. turn. At the other end, secondary leads and "out filament." Leads must not be brought out where the core comes against the end of the coil. The Smoothing Choke, HE dimensions of the stalloy for the choke core are given in a diagram. With the small current traversing its windings, this choke could be made without a gap for grid-bias purposes, but not for amplifier use. For a choke with gap, there are four sizes of stalloy, one pile of each equal to ‘the thickness of the core, in this case one inch, or 128 pieces of each size. A spool is required, with ends 2in. square, outside length, 1in. Upon this are wound 2700 turns of 36’s s.w.g. enamelled wire. Clamps are provided ag for the transformer. The radio-frequency choke consists of 1000 turns of 36’s wound upon a flat spool of two circles of stout cardboard 1}in. in diameter, separated 8-16in. by a short piece sawn off an ebonite leadin tube, or similar centre sawn from a. wooden spool. This choke is shown bolted to a strip of ebonite which in turn is screwed to the ends of the clamps of the choke.

Smoothing Condensers. ‘OUR condensers of 2 mfds. each, or two of 4 mfds. each, will be re- | quired. For bias purposes these may be of low test, say 250 volts, but for amplifier, 400 test, which will be good in either case.

_-- If the amplifier is to provide a second _QGutput voltage, an extra condenser of 2 mfds. may be required across this. e cases should be connected to earth. separate filament and secondary windings to be put on one transformer, employing separate rectifying valve and chokes for the bias and amplifier B supply. This would be a serviceable combination to run a fair-sized valve in the last stage. Reetifying Valves.

ANY small power valve that has ceased to function’ as such owing to lost filament emission, may be used as the rectifier for bias purposes. For amplifier work a "dud" power valve of not more than 5000 ohms impedance ‘should be used, as with plate and ,grid connected together, such a -valve will pass from 20 to 30 mills., and with a robust filament such as that in the quarter-ampere class, will fune_tion ‘well for a year or two if the filament voltage is not overdone. A small portion of resistance wire may be included in one filament lead for regulation purposes. General. "THD drawing shows a transformer as made from stalloy stampings, as mentioned in this column in connetion with audio chokes. If stalloy strips are used, the transformer. will stand with the coil upright in the same way as the choke. Note that the bottom of the case projects at least 7 inch on all sides, so that on three sides the tin cover may rest upon it. The flexible cord connecting with the mains enters at the back of case, over ‘Condensers, up through baseboard, conriecting to back end of fuses. A notch is cut in the corner to clear these leads, the edges being turned in to prevent. cutting. The unit will be found highly satisfactory in operation, and the precise regulation of voltage afforded will be readily appreciated by users. The writer knows that some constructors have placed upon their "B" elimin‘ator transformers a bias winding to give 100 volts. Where this is not being used, and it is desired to instal the above system, this can be done. In order not to drop more voltage than necessary in the rectifier, a valve of not more than 5000 ohms impedance should be psed. It is understood that this unit should be capable of providing a full 40 volts for a 171 valve if required at any time, and a 100-volt winding should do this. The resistances should be 50,000 ms each. ould any constructor wish to make uté\a transformer: from Ballinger’s s pings for other purposes, 8.15 turns per volt should be taken in reckoning owes —

secondary windings. This gives 2030 turns for 250 volts, 2440 for 300, and 2850 for 350 volta. If hum is heard on putting the unit into use, reverse the mains connection by turning the adapter round in the socket. The Case. "THE case is made of 3-8in. rimu, the dimensions being given in a diagram, The size of the case should be

sufficient to contain the required condensers in the lower portion, which has a door hinged at one end by a brad driven in at top and bottom. The three leads ©1, C2, O positive, are threaded before the condensers are put in place. Not less than }-inch space must be allowed above condensers, for running leads. The height of condenser space in the original is 2 1-8 inches. The panel, which may be of- any material, metal, ebonite, or 3-ply, is 54 to 6 inches high. Black sapolin all over gives suitable finish. If ‘esired, a tin cover may be pro vided, as shown, with hinged front 5 shorter tham sides, to. clear output

leads. A cover is good protection against dust, and should be connected to earth, and will also limit the magnetic tield of the transformer. Bias Resistances. HESE are variable resistances of 100,000 ohms each. If they are not purchased in potentiometer form, they should be altered so that the two ends of the resistance have each a terminal, and the arm a third terminal, for which many resistances have a hole provided, in which a bolt may be placed. One resistance is provided for each voltage required, one end of each being connected to "C" positive and the other end to "OC" negative. The arm connects to the corresponding output cable. Hach resistance gives the same range of bias voltages. ‘The only current flowing through the resistances is the "waste" current of not more than 1 mill. through each, and this they will carry, though the rating of these resistances is very low. Smoothing condengers are not required across the out-

puts. ' Connections for Grid-Bias., (THE usual half-wave eliminator ccnnections are first macs as shown on the diagram-one side of secondary to grid and plate of rectifying valve. suitable filament voltages on filament; centre-tap of filament to radio choke, then to main choke and 4 mfds., main choke to other 4 mfds. Then to one side of each resistance, cases of condensers, and "CO" positive cable. The output cables may conveniently pe of the plain rubber-covered flexible wire, which is about equal toe 20's, though much thinner wire may be used if convenient. It should here be mentioned that the choke and transformer may stand on a few thicknesses of tinfoil, which should be connected by a screw and washer holding a wire connecting to the positive lead. The negative lead from the fuse panel connects only to the remaining side of each resistance and to the remaining side of all condensers,

Connections for Amplifier Use. fey connections as given in the preceding section apply to this up to the connection. of the positive to the second 4 mfds. Then the positive lead is taken to a cable or terminal for B supply for all valvés. If a lower voltage is required for the first stage, a 200,000 ohm variable resistance may be connected to the "B" positive, and the arm will form the lower voltage output, and must also be connected to an extra 2 mfd. | condenser, of which the other side is connected to "B" negative. For detector voltage, a 100,000 variable resistance may be used as for grid-bias, as only a small current is: drawn. One side of all condensers and the .¢ases are connected together and to "B" negative cable output, which is connected to earth. Grid-bias for the amplifier is then provided by a. dry battery, but it would be quite feasible for the two